manga-translator/pipeline-translator.py

#!/usr/bin/env python3
"""
pipeline.py
───────────────────────────────────────────────────────────────
Translation OCR pipeline (No Rendering)

Usage:
  python pipeline.py /path/to/chapter/folder
"""

import os
import sys
import argparse
import zipfile
import importlib.util
from pathlib import Path

# ─────────────────────────────────────────────
#  CONFIG
# ─────────────────────────────────────────────
SOURCE_LANG = "en"
TARGET_LANG = "ca"

# Translator Settings
CONFIDENCE_THRESHOLD = 0.10
MIN_TEXT_LENGTH      = 1
GAP_PX               = "auto"
FILTER_SFX           = True
QUALITY_THRESHOLD    = 0.50
READING_MODE         = "ltr"
DEBUG                = True

# ─────────────────────────────────────────────
#  DYNAMIC MODULE LOADER
# ─────────────────────────────────────────────
def load_module(name, filepath):
    spec = importlib.util.spec_from_file_location(name, filepath)
    if spec is None or spec.loader is None:
        raise FileNotFoundError(f"Cannot load spec for {filepath}")
    module = importlib.util.module_from_spec(spec)
    spec.loader.exec_module(module)
    return module

# ─────────────────────────────────────────────
#  HELPERS
# ─────────────────────────────────────────────
def sorted_pages(chapter_dir):
    exts = {".jpg", ".jpeg", ".png", ".webp"}
    pages = [
        p for p in Path(chapter_dir).iterdir()
        if p.is_file() and p.suffix.lower() in exts
    ]
    return sorted(pages, key=lambda p: p.stem)

def make_page_workdir(chapter_dir, page_stem):
    workdir = Path(chapter_dir) / "translated" / page_stem
    workdir.mkdir(parents=True, exist_ok=True)
    return workdir

def pack_cbz(chapter_dir, translated_dir, output_cbz):
    exts = {".jpg", ".jpeg", ".png", ".webp"}

    pages = sorted(
        [p for p in Path(chapter_dir).iterdir() if p.is_file() and p.suffix.lower() in exts],
        key=lambda p: p.stem
    )

    txts = sorted(translated_dir.rglob("output.txt"), key=lambda p: p.parent.name)
    jsons = sorted(translated_dir.rglob("bubbles.json"), key=lambda p: p.parent.name)

    if not pages:
        print("⚠️  No original pages found — CBZ not created.")
        return

    with zipfile.ZipFile(output_cbz, "w", compression=zipfile.ZIP_STORED) as zf:
        # Original pages
        for img in pages:
            arcname = f"pages/{img.name}"
            zf.write(img, arcname)
            
        # Text outputs
        for txt in txts:
            arcname = f"translations/{txt.parent.name}_output.txt"
            zf.write(txt, arcname)

        # JSON outputs
        for j in jsons:
            arcname = f"data/{j.parent.name}_bubbles.json"
            zf.write(j, arcname)

    print(f"\n✅ CBZ saved → {output_cbz}")
    print(f"📦 Contains: {len(pages)} original pages, {len(txts)} text files, {len(jsons)} JSON files.")

# ─────────────────────────────────────────────
#  PER-PAGE PIPELINE
# ─────────────────────────────────────────────
def process_page(page_path, workdir, translator_module):
    print(f"\n{'─' * 70}")
    print(f"PAGE: {page_path.name}")
    print(f"{'─' * 70}")

    orig_dir = os.getcwd()
    try:
        # Isolate execution to the specific page's folder
        os.chdir(workdir)

        # 1) Translate
        translator_module.translate_manga_text(
            image_path=str(page_path.resolve()),
            source_lang=SOURCE_LANG,
            target_lang=TARGET_LANG,
            confidence_threshold=CONFIDENCE_THRESHOLD,
            min_text_length=MIN_TEXT_LENGTH,
            gap_px=GAP_PX,
            filter_sound_effects=FILTER_SFX,
            quality_threshold=QUALITY_THRESHOLD,
            export_to_file="output.txt",
            export_bubbles_to="bubbles.json",
            reading_mode=READING_MODE,
            debug=DEBUG
        )
        print("  ✅ Translator done")

        return True

    except Exception as e:
        print(f"  ❌ Failed: {e}")
        return False

    finally:
        os.chdir(orig_dir)

# ─────────────────────────────────────────────
#  MAIN
# ─────────────────────────────────────────────
def main():
    parser = argparse.ArgumentParser(description="Manga Translation OCR Pipeline")
    parser.add_argument("chapter_dir", help="Path to the folder containing manga pages")
    args = parser.parse_args()

    chapter_dir = Path(args.chapter_dir).resolve()
    output_cbz = chapter_dir.parent / f"{chapter_dir.name}_translated.cbz"

    print("Loading modules...")
    
    # Ensure we are loading from the directory where pipeline.py is located
    script_dir = Path(__file__).parent
    
    try:
        translator = load_module("manga_translator", str(script_dir / "manga-translator.py"))
    except Exception as e:
        print(f"❌ Could not load manga-translator.py: {e}")
        sys.exit(1)

    pages = sorted_pages(chapter_dir)
    if not pages:
        print(f"❌ No images found in: {chapter_dir}")
        sys.exit(1)

    print(f"\n📖 Chapter : {chapter_dir}")
    print(f"   Pages   : {len(pages)}")
    print(f"   Source  : {SOURCE_LANG} → Target: {TARGET_LANG}\n")

    translated_dir = chapter_dir / "translated"
    succeeded, failed = [], []

    for i, page_path in enumerate(pages, start=1):
        print(f"[{i}/{len(pages)}] Processing...")
        workdir = make_page_workdir(chapter_dir, page_path.stem)
        
        if process_page(page_path, workdir, translator):
            succeeded.append(page_path.name)
        else:
            failed.append(page_path.name)

    print(f"\n{'═' * 70}")
    print("PIPELINE COMPLETE")
    print(f"✅ {len(succeeded)} page(s) succeeded")
    if failed:
        print(f"❌ {len(failed)} page(s) failed:")
        for f in failed:
            print(f"   • {f}")
    print(f"{'═' * 70}\n")

    print("Packing CBZ...")
    pack_cbz(chapter_dir, translated_dir, output_cbz)

if __name__ == "__main__":
    main()